Nvidia revealed its next generation of graphics cards, the GeForce RTX 50-series, at CES this week.
There are four new GPUs in the lineup: the Geforce RTX 5090, 5080, 5070, and 5070 Ti.
The RTX 5090 and RTX 5080 launch on January 30, while the 5070 and 5070 Ti will arrive in February.
Note that these listings are for Nvidia’s reference card models; third-party 50-series GPUs will likely roll out soon after Nvidia’s models launch.
Nvidia also confirmed RTX 50-series laptop GPUs are on the way, with the first models expected to launch in March.
This week at CES, Nvidia unveiled the GeForce RTX 50-series, their next generation of graphics cards. The Geforce RTX 5090, 5080, 5070, and 5070 Ti are the four new GPUs in the lineup.
The RTX 5090, the flagship card, costs $2,000 and, according to Nvidia, is twice as powerful as the RTX 4090. However, it needs a 1,000-watt power supply, which is a significant increase from the RTX 4090’s 450W baseline power draw. The RTX 5090 fits into smaller PC builds because it is more compact than its predecessor, despite its power requirements.
All of the cards in the lineup have a smaller form factor and better performance; according to Nvidia, the $1,000 RTX 5080 is twice as quick as the RTX 4080. At almost a third of the cost, the $550 RTX 5070 provides performance on par with the RTX 4090. The 5070 Ti, meanwhile, sits between the 5070 and 5080 in terms of price ($750) and performance.
To meet these benchmarks, the new 50-series cards use Nvidia’s most recent machine learning technology, such as RTX Neural Shaders and the most recent DLSS 4 upscaling, rather than just processing power. In addition to upscaling lower-resolution gameplay to 1440p or 4K resolutions, DLSS 4 features Multi Frame Generation, which according to Nvidia can produce up to three extra frames for every rendered frame of gameplay. In essence, it gives the impression of higher, smoother frame rates in games, just like frame smoothing does in movies and TV shows, without the hardware actually rendering them all.
Since you don’t need as much horsepower to achieve high-end results, the 50-series GPUs are smaller and more affordable than their previous-generation counterparts, even though their actual hardware performance uplift is technically lower in terms of raw processing power without the aid of AI technology (the 5070, for example, has less processing power than the 4090).
When compared to rendering everything with hardware, AI upscaling and frame generation inevitably have trade-offs. Although the quantity of performance that these machine learning technologies can provide may be comparable to that of hardware, the quality of that performance is the problem. Frame generation can cause noticeable input delay, which is something that playing at high frame rates is meant to reduce rather than exacerbate. Upscalers such as DLSS, AMD’s FSR, and Sony’s PSSR on the PS5 Pro can also be inconsistent (or even incorrect, in extreme cases).
These are, to be fair, problems that Nvidia and game developers are aware of and actively attempting to resolve. The 40-series cards used DLSS to achieve significant performance gains over the 30-series, but they were also very costly. In addition, many games already make excellent use of the features. The mid-range RTX 50 GPUs may provide a far more cost-effective choice for reaching 4K resolution gameplay and high frames per second, while the 5090 could enable yet another enormous performance boost if the next generation of technology performs as planned.
GeForce RTX 5090, 5080, 5070 Ti, and 5070 preorder locations.
Keep in mind that these listings are for Nvidia’s reference card models; third-party GPUs from the 50-series will probably be released shortly after the release of Nvidia’s models. The first models of Nvidia’s RTX 50-series laptop GPUs are anticipated to go on sale in March.
Preorders for the GeForce RTX 50 series GPU from Nvidia.
